1) The document reports the discovery of 2012 VP113, the second known object after Sedna with an extremely distant perihelion of 80 AU, confirming that Sedna is not isolated. 2) Simulation results suggest there may be few inner Oort cloud objects with perihelia between 50-75 AU, and the population increases with perihelion distances greater than 75 AU. 3) Both Sedna and 2012 VP113 have similar arguments of perihelion, as do all other objects beyond 150 AU, suggesting they were perturbed by a possible unseen super-Earth mass perturber in the outer solar system that could restrict their arguments of perihelion.

1. LETTER doi:10.1038/nature13156
A Sedna-like body with a perihelion of 80
astronomical units
Chadwick A. Trujillo1
* & Scott S. Sheppard2
*
The observable Solar System can be divided into three distinct
regions: the rocky terrestrial planets including the asteroids at 0.39
to 4.2 astronomical units (AU) from the Sun (where 1 AU is the mean
distance between Earth and the Sun), the gas giant planets at 5 to
30AU from the Sun, and the icy Kuiper belt objects at 30 to 50AU
fromthe Sun.The 1,000-kilometre-diameterdwarf planetSedna was
discovered ten years ago and was unique in that its closest approach
to the Sun (perihelion) is 76 AU, far greater than that of any other
Solar System body1
. Formation models indicate that Sedna could be
a link between the Kuiper belt objects and the hypothesized outer
Oort cloud at around 10,000AU from the Sun2–6
. Here we report the
presenceof a second Sedna-likeobject, 2012VP113, whose perihelion
is 80 AU. The detection of 2012VP113 confirms that Sedna is not an
isolated object; instead, both bodies may be members of the inner
Oort cloud, whose objects could outnumber all other dynamically
stable populations in the Solar System.
The inner Oort cloud objects probably formed on nearly circular
orbits, allowing them to accumulate mass efficiently6–9
, and were later
perturbed into the eccentric orbits we see today. We define an inner
Oort cloud object as a body whose orbit is not readily formed with the
known mass in the Solar System. This typically means a perihelion
greater than 50 AU (beyond the range of significant Neptune inter-
action) and a semi-major axis in the range 150 AU , a , 1,500 AU. At
above 1,500 AU objects maybeconsideredto bein the outerOort cloud,
as galactic tides start to become important in the formation process10
.
Sedna and 2012 VP113 are the clearest examples of inner Oort cloud
objects because they donotinteract significantly with any of theknown
planets owing to their extremely distant perihelia (Fig. 1 and Table 1).
There are two main models for inner Oort cloud object formation.
In the first model, planet-sized object(s) in the outer Solar System may
perturb objects from the Kuiper belt outward to inner-Oort-cloud
orbits. These planet-sized objects could either remain (unseen) in
the Solar System5,11
or have been ejected from the Solar System during
the creation of the inner Oort cloud12
. In the second model, close stellar
encounter(s) can create the inner Oort cloud objects, possibly within
the first ten million years (Myr) of the Sun’s life, when it resided within
its birth cluster1,3,4,11,13–17
. The outer Oort cloud objects could be created
from galactic tides18
, but the more tightly bound inner Oort cloud
objectsaredifficulttocreate without aclosestellarencounter19
.A third,
less explored, model is that the inner Oort cloud objects are captured
extrasolar planetesimals lost from other stars that were in the Sun’s
birth cluster2,3,20
. Each theory of inner Oort cloud object formation pre-
dicts different orbital configurations for the population. Therefore, as
more inner Oort cloud objects are discovered, their orbits will provide
strong constraints on the inner Oort cloud object formation models
and thus our Solar System’s evolution.
To place constraints on the inner Oort cloud object population, we
constructed a simple simulation of the observational biases affecting
our survey(Methods). Wefindthree basic results fromthisanalysis: (1)
there appears to be a paucity of inner Oort cloud objects with perihelia
in the range 50–75 AU, suggesting that the inner Oort cloud object popu-
lation could increase with q . 75 AU; (2) the existence of 2012 VP113
means that the inner Oort cloud object population must reach down to
fairly small semi-major axes of about 250 AU; and (3) there are no obser-
vational biases that can explain the clustering of the argument of peri-
helion (v) near 340u for inner Oort cloud objects and all objects with
semi-majoraxesgreaterthan150 AU andperiheliagreaterthanNeptune.
Although our survey was sensitive to objects from 50 AU to beyond
300 AU, no objects were found with perihelion distances between 50 AU
and 75 AU, where objects are brightest and easiest to detect. This was
true for the original survey that found Sedna1
and the deeper follow-up
survey21
. Figure 2 illustrates this—most simulated objects (and the real
objectsSednaand2012 VP113)arefoundneartheinneredgeofthepopu-
lation.IftheinnerOortcloudobjectshadaminimumperihelionof50AU
and followed a size distribution like that of the large end of all known
*These authors contributed equally to this work.
1
Gemini Observatory, 670 North A‘ohoku Place, Hilo, Hawaii 96720, USA. 2
Department of Terrestrial Magnetism, Carnegie Institution for Science, 5241 Broad Branch Road NW, Washington, DC 20015,
USA.
Sedna
Perihelion distance (AU)
4020
0.0
0.2
0.4
0.6
0.8
1.0
8060
Eccentricity
2012 VP113
Figure 1 | Sedna and 2012 VP113 are clear dynamical outliers in the Solar
System. Eccentricity versus perihelion distance for the approximately 1,000
minor planets with well-determined (multi-year) orbits beyond 10 AU are
depicted. There are no known objects with closest-approach distances
(perihelia, q) between 55 AU and 75 AU, even though such objects would be
closer, and should therefore be brighter and easier to detect than Sedna or
2012 VP113. This suggests there may be a paucity of inner Oort cloud objects
with q , 75 AU. The perihelia of Sedna and 2012 VP113 are much too distant
from Neptune (30 AU) for their existence to be explained by the known mass in
the Solar System. All error bars are smaller than the data symbols.
2 7 M A R C H 2 0 1 4 | V O L 5 0 7 | N A T U R E | 4 7 1
Macmillan Publishers Limited. All rights reserved©2014

2. small-body reservoir distributions22,23
(q9 5 5), there would be only a 1%
chance of finding 2012VP113 and Sedna with perihelion greater than
75 AU and no objects with perihelion less than 75 AU. Therefore, we
conclude that there are few (although probably not zero) inner Oort
cloud objects in the 50–75 AU region. Some stellar encounter models
that include the capture of extrasolar material predict a strong inner
edge to the perihelion distribution of objects, which is consistent with
our observations2
.
Our simulation suggests that the inner Oort cloud contains about
1/80th of an Earth mass of material for our favoured q9 5 5 model, al-
though this can vary based on the assumed size distribution (Methods
and Extended Data Table 1). This is similar to that of the Kuiper belt
objects(KBOs),whichhaveabout1/100thofanEarthmassofmaterial22
.
We estimate that about 430z400
{240 1sð Þ inner Oort cloud objects brighter
than r 5 24.3 remain undiscovered throughout the sky. In addition,
900z800
{500 1sð Þ bodies are expected with diameter larger than 1,000 km
(most of which are extremely distant), which is statistically consistent
with detailed model estimates based only on the discovery of Sedna21
.
Our simulations assume that the inner Oort cloud objects are longitu-
dinally symmetric, but it is interesting that both 2012 VP113 and Sedna
were found at similar sky locations. If further inner Oort cloud object
discoveries prove this effect to be statistically significant, this would
strongly constrain inner Oort cloud object formation models.
Both Sedna and 2012 VP113 have similar arguments of perihelion
(v 5 311u and 293u, respectively). The orbital element v describes the
angle between the point of perihelion and where a body’s orbit crosses
thecelestialplane.Surprisingly,thisvsimilarity issharedforallknown
objects with semi-major axes greater than 150 AU and perihelia greater
than that of Neptune (v < 340u 6 55u; see Fig. 3, Extended Data Fig. 1
and Extended Data Table 2). These extreme scattered disk objects may
have a similar origin to Sedna or 2012VP113 but their much lower peri-
helia (q ,49AU) allow their existence to be explained through resonant
interactions with Neptune, except for 2004VN112, 2010GB174 and pos-
sibly2000 CR105 (refs10,24–27).Wefindthatthisvclusteringcannotbe
due to observational bias for two reasons: (1) any bias for v 50u is the
same as the bias for v 5 180u; and (2) the surveys that have found the
objects with clustered v were often off-eclipticorall-sky surveys, which
wouldnothaveabiasforeitherv 50uor180u(Methods).Thus,wecon-
clude that the v clustering is a real effect.
The Lidov–Kozai effect is the best known dynamical mechanism for
constraining the v of a minor planet28
. This three-body interaction can
create outer Oort cloud objects with v preferentially near 0u and 180u
early in the history of the Solar System when the Sun was still among its
nascent cohort of perturbing stars29
. This cannot explain the v < 0u
trend today, because v circulates owing to the presence of the giant
planets. By numerically simulating the effect of the known mass in the
Solar System on the inner Oort cloud objects, we confirmed that inner
Oort cloudobjectsshouldhaverandomv (Methods).Thissuggeststhat
a massive outer Solar System perturber may exist and restricts v for the
inner Oort cloud objects. We numerically simulated the effect of a
super-Earth-mass body at250 AU andfoundthat v for inner Oortcloud
objects librated around 0u6 60u for billions of years (see Extended Data
Figs 2 and 3). This configuration is not unique and there are many
possibilities for such an unseen perturber. A super-Earth-mass body
at 250 AU with very low albedo would be fainter than current all-sky
survey detection limits, as would larger and more distant perturbers30
.
Semimajor axis (AU)
Perihelion(AU)
200
200
180
160
80
60
140
120
100
Perihelion(AU)
200
180
160
80
60
140
120
100
300 400 500 600 700 800
Semimajor axis (AU)
a
b
200 300 400 500 600 700 800
Figure 2 | Results from the observational bias simulation. The simulated
inner Oort cloud objects (dots) are shown, along with the simulated detections
in our survey (blue crosses) and the most sensitive all-sky survey so far, which
re-detected Sedna21
(red crosses) for two different inner Oort cloud object
models: our favoured model (a) and a comparison model (b). We have
displayed ten simulated survey realizations per survey. The true values for
Sedna and 2012 VP113 are shown as red and blue circles, respectively, both with
error bars much smaller than theplot symbol (there are no error bars associated
with all other symbols, which describe simulated data). There is one difference
between the two models, the minimum perihelion for the underlying
population (75 AU in a versus 50 AU in b). Our observational results favour the
model in a because both Sedna and 2012 VP113 were found within a few AU of
their perihelion. This indicates that the inner Oort cloud objects may have
increasing numbers with increasing distance.
Table 1 | The orbital elements of the inner Oort cloud objects
Object q (AU) a (AU) e i (deg) V (deg) v (deg) tp (JD) Epoch (JD)
2012 VP113 80 266 0.698 24.051 90.806 293 2,443,300 2,455,857.1
Sedna 76.23 542 0.8593 11.92840 144.466 311.01 2,479,197 2,456,400.5
The barycentric elements for 2012 VP113 and the heliocentric elements for Sedna (from the NASA Jet Propulsion Laboratory Horizons tool for dwarf planets http://ssd.jpl.nasa.gov/?horizons) are given as semi-
major axis (a), eccentricity (e), inclination (i), longitude of ascending node (V), argument of perihelion (v), and perihelion distance (q), which has been derived from a and e. Time of perihelion (tp) and epoch are
specified in terms of the Julian date (JD). The precision to which quantities are known is reflected in the number of significant digits. We define inner Oort cloud objects as having q . 50 AU and 150 AU , a , 1,500 AU
(see main text).
RESEARCH LETTER
4 7 2 | N A T U R E | V O L 5 0 7 | 2 7 M A R C H 2 0 1 4
Macmillan Publishers Limited. All rights reserved©2014

3. Although sucha perturber could keep the v near 0u, howobjects ini-
tially obtained v < 0u and not 180u isunknown.Stochastic events,such
as a single strong stellar encounter or rogue planet could produce such
asymmetric populations. A strong stochastic encounter could also pro-
duce the low semi-major axis of 2012VP113. It is possible that the v
clustering indicates that all q . 30 AU and a . 150 AU bodies formed by
a similar process, although bodies on elliptical orbits with q less than
approximately 45AU can beexplained byNeptune interactions. Models
ofgalactictideshavedifficultyproducinginnerOortcloudobjectshaving
low semi-major axes (like Sedna) in the absence of stellar encounters19
so theeven lowersemi-majoraxisof 2012 VP113 makestides extremely
unlikely. If the inner Oort cloud objects were created from the grav-
itational tide within our Sun’s birth cluster, the low semi-major axis of
2012 VP113 constrains the birth cluster to have been a moderate10,000
solar masses per cubic parsec or more, depending on the cluster model
assumed15,29
.The moderately red opticalcolours observedfor2012 VP113
(Methods and Extended Data Table 3) are consistent with formation
in the gas giant region and not in the ultra-red dominated classical
Kuiper belt. Assuming a moderate albedo, 2012 VP113 is 450 km in
diameter,largeenoughtobeconsidered adwarfplanetifitiscomposed
largely of ice.
METHODS SUMMARY
For this work, we conducted two main observational investigations. First, we sur-
veyed 52 square degrees of sky for new inner Oort cloud objects using the Dark
Energy Camera (DECam) attheCerroTololoInter-AmericanObservatory(CTIO)
4-m telescope. Each patch of sky was imaged three times with 1.5 h to 2 h between
images, giving us the ability to detect the motion of objects beyond 300AU when
compared to background stars. Analysis software aided our search for moving
objects. After its discovery at the CTIO 4-m telescope, we measured the position
and colour of 2012VP113 using the Magellan 6.5-m telescope, completing the data
collection portion of our project. We then created an observational bias simulation
thatsimulated our discovery processto placeconstraints onthe underlying number
andorbitaldistributionofinnerOortcloudobjects.Thisproducesorder-of-magnitude
results for the number of inner Oort cloud objects whose basic parameters are
poorly constrained because so few are known.Finally, after identifying the unusual
nature of the v distribution, we simulated the inner Oort cloud objectsand the other
minor planets with semi-major axes greater than 150 AU and perihelion greater
than 30 AU to explore the long-term behaviour of v for different hypothetical
planetarymasses in the Solar System. Thesemethods aredescribedin greater detail
in the Methods section.
Online Content Any additional Methods, Extended Data display items and Source
Data are available in the online version of the paper; references unique to these
sections appear only in the online paper.
Received 11 October 2013; accepted 10 February 2014.
1. Brown, M. E., Trujillo, C. & Rabinowitz, D. Discovery of a candidate inner Oort cloud
planetoid. Astrophys. J. 617, 645–649 (2004).
2. Morbidelli, A. & Levison, H. F. Scenarios for the origin of the orbits of the trans-
Neptunian objects 2000 CR105 and 2003 VB12 (Sedna). Astron. J. 128,
2564–2576 (2004).
3. Kenyon, S. J. & Bromley, B. C. Stellar encounters as the origin of distant Solar
System objects in highly eccentric orbits. Nature 432, 598–602 (2004).
4. Melita, M. D., Larwood, J. D. & Williams, I. P. Sculpting the outer Edgeworth Kuiper
belt: stellar encounter followed by planetary perturbations. Icarus 173, 559–573
(2005).
5. Gomes, R. S., Matese, J. J. & Lissauer, J. J. A distant planetary-mass solar
companion may have produced distant detached objects. Icarus 184, 589–601
(2006).
6. Levison, H. F. & Morbidelli, A. Models of the collisional damping scenario
for ice-giant planets and Kuiper belt formation. Icarus 189, 196–212 (2007).
7. Goldreich, P., Lithwick, Y. & Sari, R. Final stages of planet formation. Astrophys. J.
614, 497–507 (2004).
8. Batygin, K., Brown, M.E.& Betts, H.Instability-driven dynamical evolution modelof
a primordially five-planet outer Solar System. Astrophys. J. 744, L3 (2012).
9. Nesvorny´, D. & Morbidelli, A. Statistical study of the early Solar System’s instability
with four, five, and six giant planets. Astron. J. 144, 117 (2012).
10. Gomes, R. S., Gallardo, T., Ferna´ndez, J. A. & Brunini, A. On the origin of the high-
perihelion scattered disk: the role of the Kozai mechanism and mean motion
resonances. Celestial Mech. Dyn. Astron. 91, 109–129 (2005).
11. Soares, J. S. & Gomes, R. S. Comparison of forming mechanisms for
Sedna-type objects through an observational simulator. Astron. Astrophys. 553,
A110 (2013).
12. Gladman, B. & Chan, C. Production of the extended scattered disk by rogue
planets. Astrophys. J. 643, L135–L138 (2006).
13. Ida, S., Larwood, J. & Burkert, A. Evidence for early stellar encounters in the
orbital distribution of Edgeworth-Kuiper belt objects. Astrophys. J. 528, 351–356
(2000).
14. Brasser, R. A two-stage formation process for the Oort comet cloud and its
implications. Astron. Astrophys. 492, 251–255 (2008).
15. Brasser, R., Duncan, M. J., Levison, H. F., Schwamb, M. E. & Brown, M. E. Reassessing
the formation of the inner Oort cloud in an embedded star cluster. Icarus 217, 1–19
(2012).
16. Dukes, D. & Krumholz, M. R. Was the Sun born in a massive cluster? Astrophys. J.
754, 56 (2012).
17. Pfalzner,S.Earlyevolution ofthebirthclusterofthesolarsystem.Astron. Astrophys.
549, A82 (2013).
18. Adams, F. C. The birth environment of the Solar System. Annu. Rev. Astron.
Astrophys. 48, 47–85 (2010).
19. Kaib, N. A., Rosˇkar, R. & Quinn, T. Sedna and the Oort cloud around a migrating
Sun. Icarus 215, 491–507 (2011).
20. Levison,H.F.,Duncan,M.J., Brasser,R.&Kaufmann,D.E.Captureofthe Sun’sOort
cloud from stars in its birth cluster. Science 329, 187–190 (2010).
21. Schwamb, M. E., Brown, M. E., Rabinowitz, D. L. & Ragozzine, D. Properties of the
distant Kuiper belt: results from the Palomar Distant Solar System Survey.
Astrophys. J. 720, 1691–1707 (2010).
22. Petit, J.-M. et al. The Canada-France Ecliptic Plane Survey—full data release: the
orbital structure of the Kuiper belt. Astron. J. 142, 131 (2011).
23. Schwamb,M.E.,Brown,M.E.&Fraser,W.C.ThesmallnumbersoflargeKuiperbelt
objects. Astron. J. 147, 2 (2014).
24. Gladman, B. et al. Evidence for an extended scattered disk. Icarus 157, 269–279
(2002).
25. Gomes, R. S., Ferna´ndez, J. A., Gallardo, T. & Brunini, A. in The Solar System Beyond
Neptune (eds Barucci, M. A., Boehnhardt, H., Cruikshank, D. P., Morbidelli, A. &
Dotson, R.) 259–273 (2008).
26. Becker, A. C. et al. Exploring the outer Solar System with the ESSENCE Supernova
Survey. Astrophys. J. 682, L53–L56 (2008).
27. Chen, Y.-T. et al. Discovery of a new member of the inner Oort cloud from the Next
Generation Virgo Cluster Survey. Astrophys. J. 775, L8 (2013).
28. Kozai, Y. Secular perturbations of asteroids with high inclination and eccentricity.
Astron. J. 67, 591 (1962).
29. Brasser, R., Duncan, M. J. & Levison, H. F. Embedded star clusters and the
formation of the Oort cloud. Icarus 184, 59–82 (2006).
30. Sheppard,S.S.et al. A southern skyandgalacticplanesurveyfor brightKuiper belt
objects. Astron. J. 142, 98 (2011).
Acknowledgements We thank the Dark Energy Camera (DECam) team for obtaining
observations during DECam commissioning, D. Norman for scheduling the November
2012 DECam observations, and D. Norman, A. Kunder and K. Holhjem for queue
observing in November. T. Abbott and F.Valdeswere very helpful during our December
2012 DECam observations. This project used data obtained with DECam, which was
Sedna
2012 VP113
Semimajor axis (AU)
Argumentofperihelion(degrees)
200 300 400 500 600100
100
0
–100
Figure 3 | The argument of perihelion for distant objects clusters about 06.
All minor planets with perihelion greater than 30 AU as a function of semi-
major axis are shown. All bodies with semi-major axis greater than the line at
150AU show a pronounced concentration near v < 0u. Errors on these orbital
elements are much smaller than the plotted symbols. This figure appears in
histogram form in Extended Data Fig. 1.
LETTER RESEARCH
2 7 M A R C H 2 0 1 4 | V O L 5 0 7 | N A T U R E | 4 7 3
Macmillan Publishers Limited. All rights reserved©2014

4. constructed by the Dark Energy Survey collaborating institutions. Observations were in
part obtained at the Cerro Tololo Inter-American Observatory, National Optical
Astronomy Observatory, which is operated by the Association of Universities for
Research in Astronomy, under contract with the National Science Foundation. This
paper includes data gathered with the 6.5-m Magellan telescopes located at Las
CampanasObservatory,Chile.ThisresearchwasfundedbyNASAPlanetaryAstronomy
grant NNX12AG26G and has also been supported by the Gemini Observatory, which is
operated by the Association of Universities for Research in Astronomy, Inc., on behalf of
the international Gemini partnership of Argentina, Australia, Brazil, Canada, Chile and
the USA.
Author Contributions C.T. created the moving object detection program for image
analysis, developed the discovery statistic simulations, and is the principal investigator
of the NASA grant supporting the project. S.S. obtained the telescope time, planned and
performed the observations, analysed the data (including the colour measurements)
andestimatedtheinnerOortcloudobjectorbitalevolutionusingtheMercuryintegrator.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Readers are welcome to comment on the online version of the paper. Correspondence
and requests for materials should be addressed to C.T. (trujillo@gemini.edu).
RESEARCH LETTER
4 7 4 | N A T U R E | V O L 5 0 7 | 2 7 M A R C H 2 0 1 4
Macmillan Publishers Limited. All rights reserved©2014

5. METHODS
Discovery survey. In our survey, we imaged 52 square degrees of sky using
DECam31
at CTIO, which has the largest field of view (2.7 square degrees) of any
4 morlargertelescope.Thisworkcoversour surveyfromNovemberandDecember
2012, which was comprised of 15 fields with mean ecliptic latitude 212.9u and 4
fields with mean ecliptic latitude 235.5u. We estimate that our survey is sensitive to
95%ofobjectsasfaintasredmagnitudemr 5 24.3witha50%detectionefficiencyat
about mr 5 24.5. Solar System objects are found by measuring their apparent
motion relative to distant, stationary background stars. Our observational strategy
istoobserveeachfieldthreetimeswith 1.5 h to2 h betweeneachimage.Our,3.5-h
time base corresponds to about 1.6 arcseconds of motion for a body at 300 AU,
which is easily visible in the ,1-arcsecond full-width-at-half-maximum image
quality of DECam. The search for new objects in the discovery data is aided by
analysis software similar to that used in other surveys32
but adapted specifically to
the DECam data set. The analysis software identifies candidate objects in the data
set, which are verified by visual inspection of the images.
Our search methodology consists of discovery with DECam and additional fol-
low-up at the Magellan telescope for any object moving at ,2.5 arcseconds per hour
(opposition distance of .50AU or farther). Ninety trans-Neptunians between 30 AU
and 50 AU were detected, but only 2012 VP113 was found beyond 50AU and required
follow-up. 2012VP113 was discovered on 5 November 2012 UT (Universal time)
moving at 1.6arcsec per hour in right ascension. We recovered 2012 VP113 again on
10–11 March 2013 UT, 10–11 August 2013 UT and 28–30 October 2013 UT with
IMACS on theMagellan Baade telescope. We further identified 2012 VP113 in archi-
val images taken at the Canada–France–Hawaii Telescope on 22 October 2011 UT.
The orbital arc33
for 2012VP113 is over two years in duration and is securely known.
Colour measurements. The physical surface colours of inner Oort cloud objects
can yield information about which formation model is likely. The classical KBOs
(,0.2 eccentricity with semi-major axes in the range 40–45 AU) appear to be dom-
inated by ultra-red colours, a rare surface colour, probably implying that their
surfaces are rich in organic material34
. We therefore obtained the optical colours of
2012VP113 intheSloang,randibandpassesatMagellanon10and11August2013UT.
The colours were determined using typical techniques35
. The colour results show
that 2012 VP113 is of only moderately red colour (see Extended Data Table 3). A
moderate red colour is consistent with formation in the gas giant region, similar to
the presumed origin of the outer Oort cloud bodies and scattered disk and not the
ultra-reddominatedclassicalKuiperbelt34,35
.Usingourphotometry,the83-AU helio-
centric distance of 2012VP113, and assuming a moderate albedo of 15% (ref. 36), we
determine that 2012VP113 is 450km in diameter.
Observational bias simulation. Because minor planets are visible in reflected
sunlight, there is a 4th-power drop in observed flux with increasing heliocentric
distance, avery strongbias against findingdistant objects. Tomodelandultimately
correct for these biases, we constructed a simple model of our observational tech-
nique. Underlying parameters characteristic of the inner Oort cloud object orbits
are assumed, and then the position and brightness of each object at a particular
time is calculated. The model population is based on a range of size distributions22
and a Gaussian scattered KBO inclination distribution37
with a uniform eccent-
ricity distribution and a power-law semi-major axis distribution32
. The values of
the assumed parameters are detailed in Extended Data Table 1, and are used to
createthe ensemble ofsimulatedobjects.We notethat althoughwe believe we have
chosen a reasonable range of parameters for the size and orbital distribution of the
innerOort cloudobjects, thesearebasedentirelyonextrapolations fromthe KBOs.
Revisionsare likelyinthefutureifthesizeandorbitaldistributionsoftheinnerOort
cloud objects deviate significantly from that of the KBOs. Once the model popu-
lation is generated, any objects brighter than our survey sensitivity limit mr 5 24.3
and withinour distancesensitivity(from50 AU, our minimumheliocentricdistance
follow-up criterion, to beyond 300 AU, where we begin to be limited by slow object
motion) are tallied as a function of ecliptic latitude. Given the typical ecliptic
latitudes of our observed fields, this tally yields the number and orbital character-
istics of the inner Oort cloud objects we should have found if our assumed distri-
bution matches the real distribution. In our observational model, we have included
our surveyas well as the deepest near all-sky survey to date, which detected Sedna21
,
because it is an additional constraint on the inner Oort cloud objects.
Our first goal of the bias simulation is to examine the perihelion distribution of
both Sedna and 2012 VP113 because both objects were found with large perihelia
(.75 AU) in surveys that were sensitive to objects with smaller perihelia. Figure 2
and Extended Data Table 1 describe an assumed size distribution of q9 5 5 (similar
to the bright KBOs and the large bodies of other stable populations)22,23,38
; how-
ever, we also test size distributions with q9 5 3.5, 4.0 and 4.5, which are found for
smaller objects in some small-body populations39
. We find that the counting
statistics for objects in the 50–75 AU region are dependent on the size distribution
assumed. For the size distributions of q9 5 3.5, 4.0, 4.5 and 5.0, we find the prob-
ability of finding Sedna and 2012 VP113 with perihelion beyond 75 AU when the
underlying size distribution extends down to 50 AU is 17%, 7%, 5% and 1%, respec-
tively. Given that we have no knowledge of the underlying size distribution of the
inner Oort cloud objects, we cannot statisticallyrule out the possibility that the 50–
75 AU regioncouldbepopulatedwithinnerOortcloudobjects.However,itappears
to have a paucity of inner Oort cloud objects compared to q . 75 AU, especially if
the inner Oort cloud objects have a steep size distribution like that of the bright
KBOs.
Our second goal for the observational simulation is to assess whether the clus-
tering of v near 0u found for Sedna, 2012 VP113 and the objects listed in Extended
Data Table 2 could be attributed to a bias effect. An v of 0u or 180u means that the
perihelion of the orbit is within the ecliptic plane. Most eccentric bodies are found
near perihelion where they are brightest, so this would suggest that surveys
restricted to the ecliptic plane should have a bias for finding objects with v around
0u or 180u. Wecanreject this bias as an explanation ofthe observed v trend fortwo
reasons. First,anybias for v 5 0u (meaningthatatperiheliontheobject crosses the
ecliptic moving southward) is the same as the bias for v 5180u (meaning that at
perihelion the object crosses the ecliptic moving northward) because observational
discovery surveys have no bias against direction of motion at this level. Because no
objects were found with v < 180u and all had v < 0u, the v clustering is already in-
consistent with any observational effect. Second, the surveys that found the objects
with clustered v were often off-ecliptic or all-sky surveys, which would not have a
bias for v 50u or 180u. Extended Data Table 2 shows the discovery ecliptic latitude
b of the objects. In our observational simulations, we find that the v 50u or 180u
discovery bias is only significant when discovery surveys probe jbj , 10u, which is
true for only about half of the objects listed. Even when this bias is present, it should
affect v 5 0u and 180u equally, for the reasons discussed above. Thus, we know of
no observational bias which could explain why all objects with q . 30AU and
a .150 AU would have v < 340u 655u and none have 90u , v , 270u.
Dynamical simulation. We used the Mercury integrator40
to simulate the long-
term behaviour of v for the inner Oort cloud objects and objects with semi-major
axes greater than 150AU andperihelia greater than Neptune. The goal of this simu-
lationwas to attempt to explain the v clustering. The simulation shows that for the
currently known mass in the Solar System, v for all objects circulates on short and
differing timescales dependent on the semi-major axis and perihelion (for example,
1,300 Myr, 500 Myr, 100 Myr and 650 Myr for Sedna, 2012 VP113, 2000 CR105 and
2010 GB17, respectively). We illustrate an example of the v circulation behaviour in
Extended Data Fig. 2. Given that the inner Oort cloud objects and the objects listed
in Extended Data Table 2 show a preference for v near 0u but not 180u, and that the
timescale for v circulation is short (and differs between objects) compared to the
ageoftheSolarSystem,wesuggestthattheperturberconstrainingvfortheseobjects
is still present and as yet unidentified.
To test the massive perturber hypothesis as a source in constraining v, we ran
several simulations with a single body of 2–15 Earth masses in a circular low in-
clinationorbitbetween 200 AU and300AU usingtheMercury integrator. Ingeneral
it was found that the v for the inner Oort cloud objects 2012 VP113 and Sedna
routinely librated with an amplitude of about 660u around v 5 0u for billions of
years. The librating behaviour of v for 2012 VP113 with an additional body of 5
Earth masses at 210AU is illustrated in Extended Data Fig. 3, where 2012 VP113
spends only 3% of its time in the 90u , v , 270u region. Sedna spent 1% of its time
in this region. For the low-i super-Earth simulations, the lower-perihelion objects
did not librate for long periods as did 2012 VP113 and Sedna and spent 30% to 50%
of their time with 90u , v , 270u. We also ran simulations with highly inclined
Neptune mass bodies at about 1,500 AU and found most objects exhibiting short v
librations, which suggests some interesting potential future simulation prospects.
Very few dynamical simulations of inner Oort cloud object formation have con-
sidered the distribution of the argument of perihelion for inner Oort cloud objects,
and so far none have produced a population that is asymmetric when comparing
v50utov5180u.Possiblestrongstochasticeventsmightcreateasymmetricpopu-
lations, but this needs to be confirmed through future simulations. We note that the
planetconfigurationswetestedarenotuniqueandtherearelikelytobemanypossible
orbital patterns for an unseen perturber that could produce a v constraint among the
most distant bodies in the Solar System41,42
.
31. Diehla,T.TheDarkEnergySurveyCamera(DECam).Phys.Proc.(Proc.2ndInt.Conf.
on Technology and Instrumentation in Particle Physics, TIPP 2011) 37, 1332–1340
(2012).
32. Trujillo, C. A., Jewitt, D. C. & Luu, J. X. Properties of the trans-Neptunian belt:
statistics from the Canada-France-Hawaii telescope survey. Astron. J. 122,
457–473 (2001).
33. Bernstein,G.&Khushalani,B.OrbitfittinganduncertaintiesforKuiperbeltobjects.
Astron. J. 120, 3323–3332 (2000).
34. Doressoundiram, A., Boehnhardt, H., Tegler, S. C. & Trujillo, C. in The Solar System
Beyond Neptune (eds Barucci, M. A., Boehnhardt, H., Cruikshank, D. P., Morbidelli,
A. & Dotson, R.) 91–104 (2008).
LETTER RESEARCH
Macmillan Publishers Limited. All rights reserved©2014

6. 35. Sheppard, S. S. The colors of extreme outer Solar System objects. Astron. J. 139,
1394–1405 (2010).
36. Stansberry, J. et al. in The Solar System Beyond Neptune (eds Barucci, M. A.,
Boehnhardt, H., Cruikshank, D. P., Morbidelli, A. & Dotson, R.) 161–179 (2008).
37. Gulbis, A.A. S. etal.Unbiasedinclination distributions for objectsintheKuiperbelt.
Astron. J. 140, 350–369 (2010).
38. Sheppard, S. S. & Trujillo, C. A. The size distribution of the Neptune Trojans and
the missing intermediate-sized planetesimals. Astrophys. J. 723, L233–L237
(2010).
39. Petit, J.-M., Kavelaars, J. J., Gladman, B. & Loredo, T. in The Solar System Beyond
Neptune (eds Barucci, M. A., Boehnhardt, H., Cruikshank, D. P., Morbidelli, A. &
Dotson, R.) 71–87 (2008).
40. Chambers, J. E. Mercury: a Software Package for Orbital Dynamics (Astrophysics
Source Code Library, 2012).
41. Lykawka, P. S. & Mukai, T. An outer planet beyond Pluto and the origin of the trans-
Neptunian belt architecture. Astron. J. 135, 1161–1200 (2008).
42. Lykawka, P. S. & Ito, T. Terrestrial planet formation during the migration and
resonance crossings of the giant planets. Astrophys. J. 773, 65 (2013).
RESEARCH LETTER
Macmillan Publishers Limited. All rights reserved©2014

7. Extended Data Figure 1 | Histogram of v for minor planets with q . 30 AU.
This is similar to Fig. 3 but in histogram form. The bodies with a . 150 AU
are shown as a black line (multiplied by a factor of ten for clarity) and bodies
with a , 150 AU are shown as a dotted line. The two distributions differ
according to Kuiper’s test with a significance of 99.9%.
LETTER RESEARCH
Macmillan Publishers Limited. All rights reserved©2014

8. Extended Data Figure 2 | The v cycling of 2012 VP113 in the current Solar
System. We note that over the course of 500Myr, the argument of perihelion v
moves uniformly across all values. All inner Oort cloud Objects (Table 1) and
other distant objects (Extended Data Table 2) are expected to exhibit this
behaviour on differing timescales, so the observation that all are restricted to v
near 0u is inconsistent with the current dynamical environment in the Solar
System. Because these are simulated plots, there are no error bars associated
with data points.
RESEARCH LETTER
Macmillan Publishers Limited. All rights reserved©2014

9. Extended Data Figure 3 | The libration of v for 2012 VP113 with an
assumed object five times the mass of Earth at 210 AU. 2012 VP113 librates
about v 5 0u for most of theduration of theSolar System. This behaviour could
explain why the two inner Oort cloud Objects (Table 1) and all objects with
semi-major axes greater than 150AU and perihelia greater than Neptune’s
(Extended Data Table 2) have v < 0u. The choice of mass and orbit of the
perturber is not unique. Many possible distant planetary bodies can produce
the pictured Kozai resonance behaviour, but the currently known Solar System
bodies cannot. These are simulated plots, so there are no error bars associated
with data points.
LETTER RESEARCH
Macmillan Publishers Limited. All rights reserved©2014

10. Extended Data Table 1 | Model parameters for the q9 5 5 inner Oort Cloud observational bias and population study
Size distributions of q9 5 3.5, 4.0 and 4.5 were also studied, with all parameters identical except that N 5 4 3 106
, N 5 1 3 107
and N 5 4 3 107
, respectively. The population mass varies greatly with the model
assumed, M 5 5, 1/30, 1/40 and 1/80 Earth masses for q9 5 3.5, 4.0, 4.5 and 5.0.
RESEARCH LETTER
Macmillan Publishers Limited. All rights reserved©2014

11. Extended Data Table 2 | Orbital elements of extreme Solar System bodies
This includes all bodies with perihelia greater than 30 AU and semi-major axes greater than 150 AU, except for the two inner Oort cloud objects listed in Table 1. The JPL Horizons heliocentric orbital elements
are given as semi-major axis (a), eccentricity (e), inclination (i), longitude of ascending node (V), argument of perihelion (v), and perihelion distance (q), which has been derived from a and e. Time of perihelion (tp)
and epoch are specified in terms of Julian Date (JD). The precision to which quantities are known is reflected in the number of significant digits. As discussed in the text, the first three objects may have
origins similar to those of the inner Oort cloud objects, although their perihelia are far lower than that of either Sedna or 2012 VP113. We also show the discovery ecliptic latitude b of the objects because the bias for
finding bodies with v 5 0u or 180u is relatively low when |b | . 10u (Methods).
LETTER RESEARCH
Macmillan Publishers Limited. All rights reserved©2014

12. Extended Data Table 3 | Colours of 2012 VP113
These colours yield a spectral slope of 13 6 2% per 100 nm and a reduced magnitude in the R-band of
3.8 6 0.04. The reduced magnitude suggests that 2012 VP113 is about 450 km in diameter, assuming a
moderate albedo of 15%.
RESEARCH LETTER
Macmillan Publishers Limited. All rights reserved©2014